Geliştirilmiş bir web tabanlı arayüz kullanarak beyin tümörlerinin manyetik rezonans görüntülerinde derin öğrenme tabanlı modellerle otomatik sınıflandırılması

Abstract

Amaç: Primer santral sinir sistemi tümörleri (PSSST), dünyada yeni teşhis edilen kanserlerin yaklaşık %3'ünü oluşturmaktadır ve erkeklerde sıklığı daha yüksektir. Beyin tümörlerinin ve PSSST'lere bağlı ölümlerin görülme sıklığı tüm dünyada giderek artmaktadır. Son zamanlarda birçok çalışma, tıbbi görüntüleme uygulamalarında derin öğrenme algoritmaları kullanılarak geliştirilen otomatik makine öğrenimi (AutoML) algoritmalarına odaklanmıştır. Bu çalışmanın temel amacı, radyologlara destek sağlamak için beyin tümörlerinin (glioma, menenjiom hipofiz adenomları) tıbbi görüntülerinin analizinde yapay zeka tabanlı tekniklerin kullanımını göstermek, hızlı ve doğru tanı konulması için beyin tümörlerini sınıflandıran kullanıcı dostu ve ‘ücretsiz web tabanli bir yazılım geliştirmektir. Gereç ve Yöntemler: Açık kaynaklı T1 ağırlıklı manyetik rezonans beyin tümörü görüntüleri Nanfang Hastanesi, Guangzhou, Çin ve Genel Hastane, Tianjin Tıp Üniversitesinden elde edildi. Önerilen web tabanlı arayüzün ve derin öğrenme tabanlı modellerin oluşturulması için Python'un programlama dilinde kullanılan Keras / Auto-Keras kütüphanesi kullanıldı. Performans değerlendirmelerinde doğruluk, duyarlılık, özgüllük, G-ortalama, F-skor ve Matthews korelasyon katsayısı ölçümleri kullanıldı. Sonuçlar: Eğitim aşamasında veri kümesinin %80'i (2599 örnek) kullanılırken, %20'si (465 örnek) test aşamasında kullanıldı. Eğitim veri setinde beyin tümörlerinin sınıflandırılmasında tüm performans ölçütleri %98'in üzerinde sonuçlanmıştır. Benzer şekilde, test veri setinde menenjiom için duyarlılık ve MCC dışındaki tüm değerlendirme ölçütleri % 91'den yüksektir. Sonuç: Deneysel sonuçlar, önerilen yazılımın üç tip beyin tümörünü tespit etmek ve tanı koymak için kullanılabileceğini ortaya koymaktadır. Geliştirilen bu web tabanlı yazılıma hem İngilizce hem de Türkçe olarak http://biostatapps.inonu.edu.tr/BTSY/ adresinden ücretsiz olarak erişilebilir.

Keywords

• Beyin tümörleri
• Derin öğrenme yaklaşımları
• Keras/Auto-Keras
• T1 ağırlıklı manyetik rezonans görüntüleme

Automated classification of brain tumors by deep learning-based models on magnetic resonance images using a developed web-based interface

Abstract

Objective: Primary central nervous system tumors (PCNSTs) compose nearly 3% of newly diagnosed cancers worldwide and are more common in men. The incidence of brain tumors and PCNSTs-related deaths are gradually increasing all over the world. Recently, many studies have focused on automated machine learning (AutoML) algorithms which are developed using deep learning algorithms on medical imaging applications. The main purposes of this study are -to demonstrate the use of artificial intelligence-based techniques to predict medical images of different brain tumors (glioma, meningioma, pituitary adenoma) to provide technical support to radiologists, and -to develop a user-friendly and free web-based software to classify brain tumors for making quick and accurate clinical decisions. Materials and Methods: Open-sourced T1-weighted magnetic resonance brain tumor images were achieved from Nanfang Hospital, Guangzhou, China, and General Hospital, Tianjin Medical University, To construct the proposed system which web-based interface and the deep learning-based models, the Keras/Auto-Keras library, which is employed in Python's programming language, is used. Accuracy, sensitivity, specificity, G-mean, F-score, and Matthews correlation coefficient metrics were used for performance evaluations. Results: While 80% (2599 instances) of the dataset was used in the training phase, 20% (465 instances) was employed in the testing phase. All the performance metrics were higher than 98% for the classification of brain tumors on the training data set. Similarly, all the evaluation metrics were higher than 91% except for sensitivity and MCC for meningioma on the testing dataset. Conclusion: The results from the experiment reveal that the proposed software can be used to detect and diagnose three types of brain tumors. This developed web-based software can be accessed freely in both English and Turkish at http://biostatapps.inonu.edu.tr/BTSY/.

Keywords

• Brain tumors
• deep-learning strategy
• Keras/Auto-Keras
• T1-weighted magnetic resonance imaging

References

1. Louis DN, Perry A, Reifenberger G, Von Deimling A, Figarella-Branger D, Cavenee WK, et al. The 2016 World Health Organization classification of tumors of the central nervous system: a summary. Acta neuropathologica. 2016;131(6):803-20.
2. Wrensch M, Minn Y, Chew T, Bondy M, Berger MS. Epidemiology of primary brain tumors: current concepts and review of the literature. Neuro-oncology. 2002;4(4):278-99.
3. Louis DN, Perry A, Reifenberger G, Von Deimling A, Figarella-Branger D, Cavenee WK, et al. The 2016 World Health Organization classification of tumors of the central nervous system: a summary. Acta neuropathologica. 2016;131(6):803-20.
4. Wrensch M, Minn Y, Chew T, Bondy M, Berger MS. Epidemiology of primary brain tumors: current concepts and review of the literature. Neuro-oncology. 2002;4(4):278-99.
5. Hassanipour S, Namvar G, Fathalipour M, Ghorbani M, Abdzadeh E, Zafarshamspour S. The incidence of brain tumours in Iran: A systematic review and meta-analysis. Advances in Human Biology. 2019;9(1).
6. Ostrom QT, Gittleman H, Fulop J, Liu M, Blanda R, Kromer C, et al. CBTRUS statistical report: primary brain and central nervous system tumors diagnosed in the United States in 2008-2012. Neuro-oncology. 2015;17(suppl_4):iv1-iv62.
7. Crocetti E, Trama A, Stiller C, Caldarella A, Soffietti R, Jaal J, et al. Epidemiology of glial and non-glial brain tumours in Europe. European journal of cancer. 2012;48(10):1532-42.
8. Howlader N, Noone A, Krapcho M, Garshell J, Miller D, Altekruse S, et al. SEER cancer statistics review, 1975–2012. Bethesda, MD: National Cancer Institute. 2015;2015.

9. Ohgaki H, Kleihues P. Population-based studies on incidence, survival rates, and genetic alterations in astrocytic and oligodendroglial gliomas. Journal of Neuropathology & Experimental Neurology. 2005;64(6):479-89.
10. Ostrom QT, Gittleman H, Liao P, Vecchione-Koval T, Wolinsky Y, Kruchko C, et al. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2010–2014. Neuro-oncology. 2017;19(suppl_5):v1-v88.
11. Bauchet L, Ostrom QT. Epidemiology and molecular epidemiology. Neurosurgery Clinics. 2019;30(1):1-16.
12. Mohammadzadeh A, Mohammadzadeh V, Kooraki S, Sotoudeh H, Kadivar S, Shakiba M, et al. Pretreatment evaluation of glioma. Neuroimaging Clinics. 2016;26(4):567-80.
13. Ostrom QT, Gittleman H, Xu J, Kromer C, Wolinsky Y, Kruchko C, et al. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2009–2013. Neuro-oncology. 2016;18(suppl_5):v1-v75.
14. Jordan JT, Plotkin SR. Benign intracranial tumors. Neurologic clinics. 2018;36(3):501-16.
15. Ezzat S, Asa SL, Couldwell WT, Barr CE, Dodge WE, Vance ML, et al. The prevalence of pituitary adenomas: a systematic review. Cancer: Interdisciplinary International Journal of the American Cancer Society. 2004;101(3):613-9.
16. Aflorei ED, Korbonits M. Epidemiology and etiopathogenesis of pituitary adenomas. Journal of neuro-oncology. 2014;117(3):379-94.
17. Drange MR, Fram NR, Herman-Bonert V, Melmed S. Pituitary tumor registry: a novel clinical resource. The Journal of Clinical Endocrinology & Metabolism. 2000;85(1):168-74.
18. Harmon SA, Tuncer S, Sanford T, Choyke PL, Türkbey B. Artificial intelligence at the intersection of pathology and radiology in prostate cancer. Diagnostic and Interventional Radiology. 2019;25(3):183.
19. Koçak B, Durmaz EŞ, Ateş E, Kılıçkesmez Ö. Radiomics with artificial intelligence: a practical guide for beginners. Diagnostic and Interventional Radiology. 2019;25(6):485.
20. Cheng J. Brain tumor dataset. figshare Dataset. 2017.
21. Cheng J, Huang W, Cao S, Yang R, Yang W, Yun Z, et al. Enhanced performance of brain tumor classification via tumor region augmentation and partition. PloS one. 2015;10(10). Cheng J, Yang W, Huang M, Huang W, Jiang J, Zhou Y, et al. Retrieval of brain tumors by adaptive spatial pooling and fisher vector representation. PloS one. 2016;11(6). Thomas T, Vijayaraghavan AP, Emmanuel S. Introduction to Machine Learning. Machine Learning Approaches in Cyber Security Analytics: Springer; 2020. p. 17-36.
22. Nielsen MA. Neural networks and deep learning: Determination press San Francisco, CA, USA:; 2015.
23. Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, et al. Imagenet large scale visual recognition challenge. International journal of computer vision. 2015;115(3):211-52.
24. Hutter F, Kotthoff L, Vanschoren J. Automated Machine Learning: Springer; 2019.
25. Bengio Y. Gradient-based optimization of hyperparameters. Neural computation. 2000;12(8):1889-900.
26. Bergstra J, Bengio Y. Random search for hyper-parameter optimization. Journal of machine learning research. 2012;13(Feb):281-305.
27. Jin H, Song Q, Hu X, editors. Auto-keras: An efficient neural architecture search system. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining; 2019.
28. Collette A, Tocknell J, Caswell TA, Dale D, Pedersen UK, Jelenak A, et al. H5Py/H5Py: 2.2. 0. 2017.
29. ZARARSIZ G, Akyildiz HY, GÖKSÜLÜK D, Korkmaz S, ÖZTÜRK A. Statistical learning approaches in diagnosing patients with nontraumatic acute abdomen. Turkish Journal of Electrical Engineering & Computer Sciences. 2016;24(5):3685-97.
30. Sartor K. MR imaging of the brain: tumors. European Radiology. 1999;9(6):1047-54.
31. Sultan HH, Salem NM, Al-Atabany W. Multi-Classification of Brain Tumor Images Using Deep Neural Network. IEEE Access. 2019;7:69215-25.
32. Sajjad M, Khan S, Muhammad K, Wu W, Ullah A, Baik SW. Multi-grade brain tumor classification using deep CNN with extensive data augmentation. Journal of computational science. 2019;30:174-82.
33. Curnes JT. MR imaging of peripheral intracranial neoplasms: extraaxial vs intraaxial masses. Journal of computer assisted tomography. 1987;11(6):932-7.
34. Wood R, Bassett K, Foerster V, Spry C, Tong L. 1.5 Tesla Magnetic Resonance Imaging Scanners Compared with 3.0 Tesla Magnetic Resonance Imaging Scanners: Systematic Review of Clinical Effectiveness: Pilot Project. 2011.
35. Yiğit V. Manyetik Rezonans Görüntüleme Sağlık Teknolojisinin Yayılımı. Türkiye Klinikleri Sağlık Bilimleri Dergisi. 2016;1(1):38-46.
36. DERNEĞİ TR. RADYOLOJİK TETKİK YOĞUNLUĞU, TETKİK YOĞUNLUĞUNDAN KAYNAKLANAN PROBLEMLERİN ANALİZİ ve ÇÖZÜM ÖNERİLERİ 2018. Available from: https://www.turkrad.org.tr/assets/2018/Radyolojik-Tetkik-Yogunlugu-Raporu.pdf.
37. Cohan RH, Davenport MS. Productivity, Meet Burnout. Academic radiology. 2018;25(12):1513-4.
38. Shanafelt TD, Hasan O, Dyrbye LN, Sinsky C, Satele D, Sloan J, et al., editors. Changes in burnout and satisfaction with work-life balance in physicians and the general US working population between 2011 and 2014. Mayo Clinic Proceedings; 2015: Elsevier.
39. Ganeshan D, Rosenkrantz AB, Bassett Jr RL, Williams L, Lenchik L, Yang W. Burnout in Academic Radiologists in the United States. Academic radiology. 2020.
40. Han S, Shanafelt TD, Sinsky CA, Awad KM, Dyrbye LN, Fiscus LC, et al. Estimating the attributable cost of physician burnout in the United States. Annals of internal medicine. 2019;170(11):784-90.
41. El-Ghandour NM. Neurosurgical education in Egypt and Africa. Neurosurgical Focus. 2020;48(3):E12.
42. Lang K, Huang H, Lee DW, Federico V, Menzin J. National trends in advanced outpatient diagnostic imaging utilization: an analysis of the medical expenditure panel survey, 2000-2009. BMC medical imaging. 2013;13(1):40.
43. McDonald RJ, Schwartz KM, Eckel LJ, Diehn FE, Hunt CH, Bartholmai BJ, et al. The effects of changes in utilization and technological advancements of cross-sectional imaging on radiologist workload. Academic radiology. 2015;22(9):1191-8.